Lightweight, Durable, and Multifunctional Electrical
Insulation Material Systems for High Voltage Applications
E. Eugene Shin1, Daniel A. Scheiman1, and Maricela Lizcano2
1Ohio Aerospace Institute, 2NASA-GRC
AIAA/IEEE Electric Aircraft Technologies Symposium (EATS)
12 - 13 July, 2018, Cincinnati, Ohio
https://ntrs.nasa.gov/search.jsp?R=20180005559 2019-08-31T14:58:32+00:00Z
Table of Contents
• Backgrounds and Objectives
• Experimental Materials
Fabrication of Dielectric Strength Test Samples
Dielectric Strength Testing
• Results and Discussion Status of the Invention
Design and Process Optimizations
Commercial Benefit/Applicability– High Voltage Power Cable
– High Voltage High Frequency Bus Bar
• Summary and Conclusions
• Future Work Plan2
Backgrounds
• Potential Electric Propulsion Architectures by J.L. Felder, NASA-GRC
3
Benefits
‒ fewer emissions,
‒ improved fuel
economy,
‒ quieter flight,
‒ improved
efficiency and
maneuverability,
‒ reduced
maintenance
costs, improved
reliability
Lots of power
transmission lines
Backgrounds
• Lightweight, high voltage, durable, and/or high temperature insulations critically needed
for future hybrid or all electric aircrafts
Power transmission bus, wiring, inter-connects, and electric motors (e.g., slot liner)
Up to 20 - 40 kV or higher e.g., require ~ 1 mm (40 mil) thick SOA Teflon-Kapton-Teflon (TKT)
0.25 to 30 MW or higherOr 10 -13 kW/kg SP motor
DC and/or AC, 400 – 4000 Hz
50 – 500 amps or higher
180 – 240 ºC or higher
Corona PD resistant
• Current HV cable technologies
not suitable for such high altitude airplane
operations particularly due to corona PD contributors
4
Backgrounds
• Original design concept of new insulation structure, so-called multilayer functional insulation
system (MFIS), on a flat conductor such as a power transmission bus bar
• Various material types with different functionalities, particularly for dielectric strength and thermal
management. Heat dissipation may need for local environment, e.g., generators (~400 ºC)5
Initial Materials Efforts in High Voltage Hybrid Electric Propulsion (HVHEP) project under
the NASA’s Convergent Aeronautics Solutions (CAS) program (June 2015 – Sept 2017)
Backgrounds
• Kapton PI film alone, 0.38 mm thick, VB=29 kV
6
New multilayer structures, namely Micro-multilayer Multifunctional Electrical Insulation
(MMEI) system, of well-known polymer insulation films, e.g., Kapton PI and PFA as
bond layer, significantly improved dielectric breakdown voltage (VB), if well-bonded.
5*KBF/5*PFA/5*KBF: 3-layers/0.38 mm th, VB=38 kV [0.5*HPP/1*PFA]9 /0.5*HPP: 19-layers/0.38 mm th, VB=46 kV
Objectives
Under the NASA’s Transformational Tools and Technology (TTT) program (Oct 2017 - )
• To maximize dielectric performance of the new MMEI structures via material-design-process optimizations
• To incorporate multifunctionalities, such as high partial discharge resistance, improved durability, EMI shielding, and high thermal dissipation
• To demonstrate scale-up and commercial applicability of MMEI system
7
Experimental: Materials
1, 2, and 5 mil thick Thermalimide Kapton bagging film (KBF) from Airtech
International, Inc. as the baseline PI
Kapton® PI films from DuPont
‒ 0.3, 1, 5 mil thick HN, a tough, aromatic film: 30HN, 100HN, 500HN, respectively
‒ 50HPP-ST, 0.5 mil thick with superior dimensional stability and adhesion characteristics
‒ 100CRC, 1 mil thick, corona resistant films
PFA films: 0.5 and 1 mil films from Chemours; 2 and 5 mil films from McMaster-Carr
2 mil thick virgin Teflon® PTFE films from McMaster-Carr
2 mil thick PET, Mylar A polyester films from Tekra
2 mil thick thermally conductive PI (TCPI) films from McMaster-Carr
1 mil thick electrically conductive PI (ECPI) films from McMaster-Carr
1.5 mil thick eGRAF® Spreadershield™ SS1500 flexible graphite from GrafTech
8
Initial Candidate Materials, all commercially available
Experimental: Fabrication of Dielectric Strength Test Samples
9
Constituent films,
cleaned with alcohol
and air dried
Sample Batch #1
1/16” thick
aluminum sheets
Laid up (2×1.25”);
PI vs PFA for skin layers
Heated in oven under
compression w binder clips
Consolidated coupon
for 2 test samples
Sample Batch #2
3/16” thick A2
tool steel Laid up (3×1.25”)
Heated in oven under
compression w HT
sealing clips
Consolidated coupon for 3
test samples
Experimental: Fabrication of Dielectric Strength Test Samples
Optimized heat fuse-bonding conditions:
• PFA: heat to 350 ºC, dwell for 10 min (allowed to 353 ºC)
• PET: heat to 270 °C, dwell for 10 min
under uniform compression loading of about 8 to 8.4 psi using
either 14 clips# on 2 × 1.25 inch coupon or 20 clips# on 3 ×1.25 inch coupon
# Inconel high temperature sealing clip: rated to 370 °C, 1.5 lbs
clamping force per clip
10
Experimental: Dielectric Strength Testing
Commercial test rig, Model DT2-60-20-SR-P-C by Sefelec Eaton, France,
used for material screening
11
Standardized Test
Conditions
• TF3 fixture (0.25” dia.
electrodes with edges
rounded to 0.0313” R)
• Oil bath with PM-125
phenylmethylsiloxane
• Simple AC ramp at 0.6 kV/s
• Run reference samples with
known VB before and after
every actual sample group
Results & Discussions: Status of Invention
12
Overall dielectric performance of MMEI structures
• Parameters that control VB or K of MMEI: total thickness,individual layer thickness, total accumulated thicknesses of constituent materials, overall thickness ratio of constituent materials, and total number of layers or interfaces in addition to bonding integrity.
Proc'ed
mil mm mm
BS11 5*PFA/5*KBF/5*PFA 15 0.381 0.363
BS12 5*KBF/5*PFA/5*KBF 15 0.381 0.378
BS13 2*PFA/5*KBF/5*PFA/5*KBF/2*PFA 19 0.483 0.455
BS14 5*KBF/5*PFA/1*KBF/5*PFA/5*KBF 21 0.533 0.478
BS15 [2*PFA/2*KBF]3 /2*PFA 14 0.356 0.363
BS16 1*KBF/2*PFA/2*KBF/5*PFA/2*KBF/2*PFA/1*KBF 15 0.381 0.345
BS17 [1*KBF/2*PFA]4 /1*KBF 13 0.330 0.338
BS17N [1*KBF/1*PFA]4 /1*KBF 9 0.229 0.233
BS18 [0.5*PFA/1*KBF]6 /0.5*PFA 9.5 0.241 0.252
BS19 [1*KBF/0.5*PFA]4 /1*KBF 7 0.178 0.178
BS20 [0.3*HN/0.5*PFA]16 /0.3*HN 13.1 0.333 0.350
BS20S [0.3*HN/0.5*PFA]4 /0.3*HN 3.5 0.089 0.089
BS21 [0.5*HPP/0.5*PFA]9 /0.5*HPP 9.5 0.241 0.254
BS22 [0.5*HPP/1*PFA]9 /0.5*HPP 14 0.356 0.386
BS23N [1*KBF/2*PET]4 /1*KBF 13 0.330 0.158
BS23 [1*KBF/2*PET]4 /1*KBF 13 0.330 0.210
BS12 5*KBF/5*PFA/5*KBF 15 0.381 0.343
BS17NH [1*HN/ 1*PFA]4 /1*HN 9 0.229 0.225
BS17NC [1*CRC/ 1*PFA]4 /1*CRC 9 0.229 0.242
BS17NHT 2*PTFE/1*PFA/[1*HN/ 1*PFA]4 /1*HN 12 0.305 0.310
BS19 [1*KBF/0.5*PFA]4 /1*KBF 7 0.178 0.173
BS20S [0.3*HN/0.5*PFA]4 /0.3*HN 3.5 0.089 0.092
BS20SR [0.3*HN/1*PFA]4 /0.3*HN 5.5 0.140 0.149
BS20ST [0.3*HN/2*PFA]4 /0.3*HN 9.5 0.241 0.238
BS20US [0.3*HN/0.5*PFA]2 /0.3*HN 1.9 0.048 0.049
BS21 [0.5*HPP/0.5*PFA]9 /0.5*HPP 9.5 0.241 0.239
BS21S [0.5*HPP/0.5*PFA]2 /0.5*HPP 2.5 0.064 0.069
BS22 [0.5*HPP/1*PFA]9 /0.5*HPP 14 0.356 0.380
Ba
tch
#1
Coupon ID
Ba
tch
#2
Layer ConfigurationOverall Thickness
Design
Thickness limit, ~ 0.15
Highest increase: ~ 61%
86.3% thicknessreduction
* indicated thickness in mil (1/1000 inch)
× GRC in-house test results +
literature values up to 0.76 mm thick
Results & Discussions: Status of Invention
13
Effects of individual layer thickness
• For both PI and PFA, K of MMEI structures increased with
decreasing layer thickness, but leveled off at ~ 0.05 mm
which seemed to be related to the overall thickness limit.
• Both PI and PFA contributed to the overall performance of
MMEI structures.
Results & Discussions: Status of Invention
14
Effects of total constituent material thickness
• In general, K of MMEI structures increased with increasing
total thickness of PI layers, but decreased with increasing
that of PFA or PET layers.
• Contribution of PI layers on the overall K of MMEI
structures was greater than that of PFA or PET bond layers.
Results & Discussions: Status of Invention
15
Effects of total number of layers or interface
• K of MMEI structures increased slightly with increasing number of interface or total number of layers,
but decreased when their overall thickness was less than ~ 0.18 mm, which seemed to be related to
the overall thickness limit on the MMEI effectiveness.
Results & Discussions: Status of Invention
16
Effects of PI material modifications
• Various types of PI considered for multifunctionalities of MMEI
• Modifications, typically via addition of fillers or additives, decreased K of either PI alone film or MMEI
structures. Much worse in MMEI structures, especially for the corona resistant CRC PI film
• Unexpected drop of K by adding 2 mil thick PTFE layer in BS17NHT, possibly due to weaker bonding
between PFA and PTFE?
Results of all those semi-quantitative analyses are systematically applied to design
future, more efficient MMEI structures with maximized VB or K.
Results & Discussions: Status of Invention
17
Typical dielectric breakdown failure modes
• The hole size in PFA films increased with increasing thickness but only up to ~0.025 mm.
• Failure mode in PI changed with increasing thickness, i.e., melting, charring, and THP up to 0.013 mm
additional micro-cracking before THP up to 0.025 mm PP at and above ~0.05 mm.
• Note that DZW in (d) defined the width of damage band around the electrodes.
100 µm 100 µm 100 µm 1.5 mm(a) (b) (c) (d)
DZW
PFA
Kapto
n P
I
(a) 0.013 mm thick
(b) 0.025 mm thick
(c) 0.05 mm thick
(d) 0.125 mm thick;
All showed melt, char, THP (Thru-
hole perforation) or PP (partial
perforation).
(a) 0.013 mm th.; char, THP
(b) 0.025 mm th.; cracks, char, THP
(c) 0.05 mm th.; cracks, cavitation,
melt, char, PP
(d) 0.126 mm th.; extensive cracks-
cavitation, melt, char, PP
HPP-ST HN KBF HN
Results & Discussions: Status of Invention
18
Typical dielectric breakdown failure modes
• Additional unique failure mode in MMEI structures was debonding or inter-layer separations.
• Failure mode of MMEI structures affected by (i) overall thickness, (ii) individual layer thickness, (iii) PI/BL ratio.
• Most MMEI structures involved microcracking or cracking, cavitation, debonding, melting, charring, and PP,
but the ultimate breakdown occurred via the localized charring and PP.
• Samples failed by THP normally showed lower VB or K.
MM
EI
(a) 0.05 mm th. BS20US; cracks,
debond, melt, char, THP,
(b) 0.125 mm th. BS20SR,
(c) 0.173 mm th. BS19,
(d) 0.31 mm th. BS17NHT,
(e) 0.2235 mm th. BS17NH,
(f) 0.345 mm th. BS16,
(g) 0.38 mm th. BS22,
(h) 0.455 mm th. BS13;
(b) thru (h) all showing cracks,
cavitation, debond, melt, char,
and PP.
Results & Discussions: Status of Invention
19
Dielectric breakdown damage evolution in MMEI structures
• Damage evolution sequence in MMEI structures was determined experimentally as a function of V.
• From the extensive failure mode analyses, for a given overall thickness, the failure mode transitioned from
more catastrophic mode involving cracking, cavitation, charring, THP in single polymer insulation films to
more gradual or progressive mode involving microcracking, cavitation, melting, channeling,
debonding, interfacial swelling, charring and PP in the new MMEI structures.
Micro-cracking, cavitation, presumably in
Kapton PI layers at 22 kV
Extensive crack propagation, cavitation, localized debonding,
melting, charring, and PP at dielectric breakdown at 36 kV
BS17N; [1*KBF/1*PFA]4 /1*KBF; 9-layers, 0.233 mm thick
Results & Discussions: Status of Invention
20
K and DZW correlation in MMEI structures
• In MMEI structures, the higher K was, the larger DZW was or vice versa regardless of overall thickness.
• Dielectric breakdown failure of MMEI structures proceeded with a progressive damage evolution involving
more damage types/events and larger damage zones, i.e., more energy involved in the breakdown process
the higher dielectric strength.
Results & Discussions: Design and Process Optimizations
21
Example of optimum MMEI design
• In spite of significant performance improvement of the MMEI structures to date, their performance can be
further improved when their design-structural configurations, insulation material types, and process-
fabrication conditions are optimized.
• MMEI structures can incorporate multifunctionalities by the nature of their design capabilities, such as
Corona PD resistance, EMI shielding, mechanical durability, or thermal management, etc.
Results & Discussions: Design and Process Optimizations
22
Potential evidence of improved structural durability
• Brittle to ductile transition occurred with increasing PC v% or decreasing layer thickness in PC/SAN
multilayer structures.
• Ref: E.E. Shin, A. Hiltner, and E. Baer, "The Brittle-to-Ductile Transition in Microlayer Composites", J. of
Applied Polymer Science, 47, p269, 1993.
BRITTLE
1 SAN-control
2 30/70 (4/8), 194L
3 25/75 (11/34), 49L
4 53/47 (23/22), 49L
PC/San v% (layer thickness, µm),
# of layers
SEMI-BRITTLE
1 65/35 (27/15), 49L
2 45/55 (4.9/5.4), 194L
3 40/60 (1.3/2), 776L
4 74/26 (33/13), 49L
5 53/47 (3.6/3.3), 388L
DUCTILE
1 55/45 (1.5/1.2), 776L
2 76/24 (9/3), 194L
3 67/33 (2/1), 776L
4 PC-control
Results & Discussions: Commercial Benefit/Applicability
23
HV Power Pod Cable by GORE
• Unique design developed to carry 0.25 MW at 15 kV (but rated to 40 kV), and applicable to -80 C to >260 ºC
use temperature
• Consisted of six identical conductor pods insulated by the GORE’s proprietary PTFE-PTFE composite and
arranged horizontally by a corona resistant PTFE jacket
• By GORE’s standard testing, VB of the PTFE-PTFE composite insulation was ~ 39 kV, but it dropped to ~ 29
kV when the PTFE jacket was added. The cause not fully identified.
Results & Discussions: Commercial Benefit/Applicability
24
HV Power Pod Cable by GORE: VB and failure mode measurement at GRC
• The optimized bend configuration produced reproducible VB, 30.2±1.7 kV, consistent with the GORE data.
• VB of the cable after removing jacket was also 29.9±0.04 kV, which suggested that the jacketing process at
GORE seemed to cause permanent change in the main PTFE composite insulation.
• 600V silicone hookup wire performed exceptionally well, reached 23.9 kV breakdown voltage.
• Determination of Inter-conductor VB was not successful for the GORE cable.
Results & Discussions: Commercial Benefit/Applicability
25
HV Power Pod Cable by GORE: Design options to apply MMEI
• Either a modified BS22 ([0.5*HPP/1*PFA]9 /0.5*HPP, 0.38 mm th) or BS17NH ([1*HN/1*PFA]4 /1*HN, 0.225 mm th)
MMEI configuration was considered.
• Option II: ideally more solid insulation VS. Option I: simpler & easier to process, e.g., vacuum-bagging/autoclaving
• Overwrap seamlines in option II can be a weak spot for high voltage due to trapped air.
Results & Discussions: Commercial Benefit/Applicability
26
HV Power Pod Cable by GORE: Design optimization of Option I
• Experimental determination of web width (= inter-conductor spacing - 2×MMEI thickness) via
measurement of VB through PFA bondline or PFA-PI interface
• Four different inter-conductor spacing tested: 1, 2, 3, and 5 mm; 4 repeats per spacing
• Successful dielectric strength testing in the GRC test rig Failure mode analysis
Results & Discussions: Commercial Benefit/Applicability
27
HV Power Pod Cable by GORE: Design optimization of Option I
• Validation of failure path through PFA bondline or PI-PFA interface, thus validated test results
PB1-5, 1 mm spacing: melt, cavitation, possible
perforation on PFA, PFA-PI
interface debond, cracks on PI
PB5-7, 5 mm spacing: melt, cavitation, &
possible perforation on
PFA; PFA-PI interface
debond
• 5 mm spacing selected for the 40 kV requirement including a safety factor consideration
• Final OD of option I or II cable may vary depending on modification of MMEI configuration to
incorporate other multifunctionalities, but yet significantly thinner than that of GORE cable
• Initially, ~ 1 m long cables to be fabricated
Results & Discussions: Commercial Benefit/Applicability
28
HV Power Pod Cable by GORE: Design optimization of Option I
TH, mm OD, mm TH, mm OD, mm TH, mm OD, mm
Conductor: AWG 10 (37/0.404mm),
Ni plated copper2.70 2.70 2.70
PTFE-PTFE Composite 1.34 5.38
MMEI: modified BS22 0.394 3.49
MMEI: modified BS22 0.381 3.46
PTFE, Corona resistant 0.3 5.98
5 mil PFA + 5 mil PTFE 0.254 3.97
Web width: 2 4.2 3
Web thickness: 0.6 0.787 0.508
Final Cable 5.98 3.49 3.97
Overall width, mm
Insulation:
Jacket:
51.2 41.8
GORE
53.5
Flat cable with 6 high voltage
conductors AWG 4 equivalent
MMEI Option I MMEI Option II
Results & Discussions: Commercial Benefit/Applicability
29
HV high frequency bus bar with MERSEN
• A three-phase system for 1 MW up to 10 MW operating power with operating voltage of 20 kV (
designed for 40 kV), high frequency (400 Hz up to 4000 Hz), and temperature up to 180 ºC
• Three prototypes with three different conductor thickness, fully insulated by MERSEN
• Two sets of blank conductors to apply MMEI structures
Summary and Conclusions
• Multilayer structures of well-known polymer insulation materials, namely MMEI, were
newly developed and evaluted for HV insulation. Based on extensive evaluations to
date, key findings are as follows:
‒ MMEI structures with various Kapton PI materials and PFA or PET as a bond layer
achieved 61% increase in VB or K compared to that of Kapton PI alone films or the SOA
TKT, thus resulted in 86.3 % decrease in insulation thickness.
‒ Dielectric performance of MMEI structures was governed by various material, process,
and structural parameters, such as dielectric properties of constituent materials, inter-layer
bonding integrity, overall thickness, total number of layers or interface, individual layer
thickness, and ratio of constituent materials.
‒ Good inter-layer bonding integrity was essential for improved VB or K.
‒ K of the MMEI structures increased with (i) decreasing individual layer thickness
regardless of material type, (ii) increasing total accumulated thickness of PI or overall
PI/BL ratio, and (iii) increasing number of interface or total number of layers, but only
above the overall thickness limit of 0.15 mm. 30
Summary and Conclusions, Cont’d
31
‒ Contribution of Kapton PI on overall MMEI dielectric performance was greater than that of
PFA or PET, and as a bond layer PFA performed better than PET.
‒ For a given overall thickness, the failure mode seemed to change from more catastrophic
mode involving cracking, cavitation, charring, PP or THP in single polymer insulation films
to more gradual or progressive mode involving microcracking, cavitation, melting,
channeling, debonding, interfacial swelling, charring and PP in the new MMEI structures.
‒ Dielectric breakdown failure of MMEI structures proceeded with a progressive damage
evolution involving more damage types/events and larger damage zones, which
suggested that more energy was involved in the breakdown process, thus resulted in the
higher dielectric strength.
‒ Material modifications, typically via addition of fillers or additives, decreased K in either PI
alone film or MMEI structures since the fillers or additives, especially their interfaces with
matrix material, acted as defects.
‒ Various responsible mechanisms for the significant property improvement of the new
MMEI system were postulated, but should be validated experimentally.
Summary and Conclusions, Cont’d
32
‒ Improvement of processing, e.g., more accurate control of fuse-bonding temp, compression
loading at all temp, and cleanliness, granted additional increase of VB in various MMEI
structures, but thinner structures below the limitation, 0.15 mm, was less affected.
• Design and performance evaluations of scaled-up MMEI system were initiated to
validate their practicality and applicability using the SOA commercial HV power
transmission systems including GORE’s HV power pod cable and MERSEN’s HVHF
bus bar prototypes:
‒ VB of GORE cable was measured successfully using the GRC dielectric strength test rig.
‒ Two options to apply MMEI system to GORE pod cable were developed including
determination of optimum dimensions, fabrication methods and procedures.
‒ With MERSEN, a meter-long 3-phase bus bar prototype has been developed for 1 MW up
to 10 MW power with operating voltage of 20 kV ( designed for 40 kV), high frequency (400
Hz up to 4000 Hz), and up to 180 ºC use temp.
‒ Design and fabrication procedures for applying MMEI system to the same blank bus bar are
being developed.
Future Work Plan
The following tasks are planned to continue for development
and improvement of the MMEI system:
• Material-design-process optimizations, especially for
multifunctionalities including inorganics, ceramics, or metals
• Scale up and commercialization feasibility assessment
• More sophisticated performance evaluations of the MMEI
structures including synergistic durability assessment
• Experimental validation of potential mechanisms on
performance enhancement of MMEI structures
33
Acknowledgments
W. L. GORE & ASSOCIATES, INC., Landenberg, PA
MERSEN New Product Development, Rochester, NY.
Special thanks to A. Woodworth, Janet Hurst, and the rest of project team at GRC.
This work has been sponsored by NASA’s Convergent Aeronautics Solutions (CAS) program initially, and by Transformational Tools and Technology (TTT) program currently, as a part of NASA’s Transformative Aeronautics Concept Program (TACP) under Aeronautics Research Mission Directorate (ARMD).
34
Thank You for your attention!
Any Questions?
Top Related